University of Wisconsin MilwaukeeUWM Digital Commons

Theses and Dissertations

December 2016

Corrosion of Aluminum Current Collector in CostEffective Rechargeable Lithium-Ion BatteriesShengyi LiUniversity of Wisconsin-Milwaukee

Follow this and additional works at: https://dc.uwm.edu/etdPart of the Chemistry Commons, Materials Science and Engineering Commons, and the Oil,

Gas, and Energy Commons

This Dissertation is brought to you for free and open access by UWM Digital Commons. It has been accepted for inclusion in Theses and Dissertationsby an authorized administrator of UWM Digital Commons. For more information, please contact open-access@uwm.edu.

Recommended CitationLi, Shengyi, "Corrosion of Aluminum Current Collector in Cost Effective Rechargeable Lithium-Ion Batteries" (2016). Theses andDissertations. 1384.https://dc.uwm.edu/etd/1384

https://dc.uwm.edu/?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttps://dc.uwm.edu/etd?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttps://dc.uwm.edu/etd?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/131?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/285?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/171?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/171?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPageshttps://dc.uwm.edu/etd/1384?utm_source=dc.uwm.edu%2Fetd%2F1384&utm_medium=PDF&utm_campaign=PDFCoverPagesmailto:open-access@uwm.edu

CORROSION OF ALUMINUM CURRENT COLLECTOR IN

COST EFFECTIVE RECHARGEABLE LITHIUM-ION

BATTERIES

by

Shengyi Li

A Dissertation Submitted in

Partial fulfillment of the

Requirements for the Degree of

Doctor of Philosophy

in Engineering

at

University of Wisconsin-Milwaukee

December 2016

ii

ABSTRACT

CORROSION OF ALUMINUM CURRENT COLLECTOR IN COST

EFFECTIVE RECHARGEABLE LITHIUM-ION BATTERIES

by Shengyi Li

The University of Wisconsin-Milwaukee

Under the Supervision of Professor Benjamin Church

Rechargeable lithium ion batteries (LIB) have been widely used as commercial energy

storage systems for portable equipment, electronic devices and high power applications (e.g.

electronic vehicles). One issue with the commercialized LIB is that expensive, highly toxic and

flammable organic solvents are used in the electrolyte and the fabrication process of electrodes.

The toxic organic based solvents increase the production cost and lead to significant safety

concerns in the event of a battery overcharge or short circuit. The recent development of “green

manufacturing” technology allows manufacturers to replace the organic solvents used in the

cathode coating process by aqueous based slurries. In addition, the further transition from

organic based LIB system to completely aqueous based lithium ion battery (ARLB) has attracted

a lot of attention recently because of its potential to significantly reduce manufacturing cost and

eliminate the risks and environmental issues associated with the commercialized, organic based

lithium ion batteries. Such new aqueous-based technologies often use basic aqueous solutions

with high pH value, which brings concerns on the possible occurrence of aluminum current

iii

collector corrosion. The corrosion of aluminum current collector in lithium ion batteries is one of

the possible factors that affect the long-term performance and safety of lithium-ion batteries. In

this work, the corrosion phenomenon of aluminum current collector in lithium ion batteries that

use aqueous-based chemistries is explored experimentally and theoretically. Here, the corrosive

aqueous media defined in lithium-ion battery systems includes the aqueous based slurry used in

the fabrication of cathode coating, aqueous lithium nitrate electrolyte and aqueous lithium sulfate

electrolyte. This research aims to reveal the corrosion behavior, corrosion mechanisms and

corrosion kinetics of aluminum in exposure to aqueous environment during the fabrication and

service life of aqueous-based lithium-ion battery systems, and shed light on the management of

corrosion in the design of cost effective lithium ion batteries.

Corrosion of aluminum can occur during the manufacturing of lithium ion batteries when

aqueous-based cathode slurries is used during cathode coating process. The corrosion mechanism

of AA1085 in exposure to aqueous based cathode slurry was investigated by surface

characterization on aluminum after exposure tests and measuring electrochemical characteristics.

In exposure tests, the alkaline pH value of aqueous-based cathode slurries and immersing time

were revealed as the principle factors that control the corrosion of aluminum during the cathode

manufacturing process. The nickel manganese cobalt oxide active material used in the slurry

does not have a direct impact on corrosion of the aluminum current collector. The initiation and

evolution of localized corrosion on aluminum are closely related to the formation of galvanic

cells between aluminum matrix and intermetallic particles. X-ray photoelectron spectroscopy

iv

confirmed that the pH of cathode slurry was the only factor that influence the surface

composition of aluminum. The oxide passive film gradually degraded into hydroxide with the

elapsing exposure time. Electrochemical characterizations showed that aluminum electrodes

gave remarkably different response to the different pH of test solutions. The time-pH-variant

electrochemical response was ascribed to the change of passive film and electric double layer

properties.

The electrochemical stability of high-purity aluminum in 2 M Li2SO4 and 5 M LiNO3

ARLB electrolytes was evaluated over a range of pH conditions by cyclic voltammetry, linear

sweep voltammetry and chronoamperometry. Aluminum presented high corrosion resistance at

pH 5, pH 7 and pH 9 within the stability windows of both electrolytes. At the pH 11 condition, 2

M Li2SO4 is capable of inhibiting aluminum from pitting, although the inhibiting effect is not

sustainable and crystallographic pitting occurs under a continuously applied anodic potential.

Aluminum was well passivated against pitting in 5 M LiNO3 electrolyte at pH 11 due to the

formation of a thick corrosion product barrier layer. Raman spectra showed the presence of

sulfate and nitrate anions on aluminum surface after cyclic voltammetry at pH 11. The chemical

adsorption mechanisms of sulfate and nitrate anions on aluminum were proposed to explain the

dependency of electrochemical stability of aluminum on pH, anodic potential and type of anions.

The applicability of aluminum as current collector in ARLB using the 2 M Li2SO4 and 5 M

LiNO3 electrolytes was discussed.

v

The corrosion kinetics of AA1085 in Li2SO4 and LiNO3 aqueous rechargeable lithium-ion

battery electrolytes at pH 11 under the influence of various experimental variables was studied

using chromoamperometry. AA1085 is susceptible to crystallographic pitting corrosion in

Li2SO4 electrolytes. The rate of pit nucleation and the rate of pitting growth on AA1085 both

decreased at higher Li2SO4 concentrations or at lower anodic potentials. In LiNO3 electrolytes,

AA1085 was passivated against pitting corrosion due to the formation of a thick, uniform

corrosion product layer. The repassivation rate was slightly enhanced by increasing the

electrolyte concentration and anodic potentials. X-ray photon electron spectroscopy spectra

showed the formation of a thin sulfate-incorporated passive film, which comprises

Al2(SO)418H2O, Al(OH)SO4 and Al(OH)3 on electrode before the occurrence of pitting growth

in 2 M Li2SO4 electrolyte. The thick corrosion product layer formed in 5 M LiNO3 electrolyte is

composed of Al(OH)3 and AlOOH. Raman spectroscopy on deionized water, LiOH solution,

Li2SO4 and LiNO3 depicted changes of solution structure with increasing electrolyte

concentrations. The influence of extrinsic factors, including the alkaline solution and the anodic

potential, and intrinsic factors, such as the surface chemical adsorption of anions, chemical state

of passive films and dissolubility of electrolytes, on the corrosion kinetics of AA1085 in slightly

alkaline Li2SO4 and LiNO3 electrolytes are revealed.

The intermetallic particles containing Fe and Si in aluminum alloys have electrochemical

potentials that differ from that of aluminum matrix, resulting in the formation of galvanic

couples and detrimental pitting corrosion. The electrochemical characteristics of AA1100,

vi

surface treated AA1100 with “intermetallic-free” surface, home-synthesized Al2Fe and

Al2FeSi0.67 alloy were measured by potentiodynamic polarization in alkaline solutions with the

addition of Li2SO4 and LiNO3. In general, intermetallic alloys presented noble corrosion

potentials compared to AA1100 specimens. The addition of sulfate anions in the solution does

not suppress the selective dissolution of aluminum on intermetallic alloys in 0.001 M and 1 M

LiOH solutions, which increases the cathodic efficiency of intermetallic alloys and promotes the

galvanic corrosion. The corrosion potential difference is significantly reduced when 2 M LiNO3

is added into the alkaline solution. Meanwhile the anodic dissolution rate that corresponds to the

preferable dissolution of Al also decreases. Raman spectra revealed that the inhibiting effect of

LiNO3 on selective dissolution of aluminum is due to the formation of Fe3O4 passive film above

the corrosion potential. the cathodic polarization curves showed that the intermetallic alloys

sustain higher cathodic current than AA1100 and surface-treated Al. The magnitude of cathodic

current density measured on the electrodes follows the following order:

Al2Fe>Al2FeSi0.67>AA1100>surface-treated AA1100. The change of composition and structure

on the intermetallic surface during anodic polarization influences the selective dissolution

process, the passivity status and in turn affects the cathodic efficiency of the intermetallic.

vii

© Copyright by Shengyi Li, 2016

All Rights Reserved

viii

TABLE OF CONTENTS

LIST OF FIGURES .................................................................................................................... xii

LIST OF TABLES ...................................................................................................................... xvi

LIST OF NOMENCLATURE ................................................................................................. xvii

ACKNOWLEDGEMENT ....................................................................................................... xviii

CHAPTER 1 Introduction ........................................................................................................... 1

1.1 Background and motivation ............................................................................................................ 1

1.2 The role of structural and compositional features on Al corrosion ............................................. 3

1.2.1 The protective surface passive film ............................................................................................. 3

1.2.2 The role of aluminum intermetallic particles .............................................................................. 5

1.3 Aqueous based lithium-ion battery system .................................................................................... 9

1.3.1 Advantages of aqueous based lithium ion battery system ........................................................... 9

1.3.2 Stability window of aqueous electrolytes aqueous based lithium ion battery ........................... 12

1.3.3 Recent progress on development of aqueous based lithium ion batteries ................................. 14

1.4 Corrosion of aluminum current collector in lithium ion battery ............................................... 15

1.4.1 Requirements for current collectors in lithium-ion battery ....................................................... 15

1.4.2 Factors that influence aluminum corrosion in non-aqueous based lithium-ion battery ............ 19

1.4.3 Adverse effect of corrosion of current collector on lithium-ion battery performance ............... 23

1.5 Methods for protecting aluminum current collector from corrosion ........................................ 25

1.5.1 Inhibitor ..................................................................................................................................... 25

1.5.2 Growth of oxide passive film .................................................................................................... 26

1.5.3 Protective coating ...................................................................................................................... 28

CHAPTER 2 Effect of Aqueous-based Cathode Slurry pH and Immersion Time on

Corrosion of Aluminum Current Collector ................................................................ 30

2.1 Introduction .................................................................................................................................... 30

2.2 Experimental .................................................................................................................................. 32

ix

2.3 Results and analysis ....................................................................................................................... 35

2.3.1 Immersion test ........................................................................................................................... 35

2.3.2 Electrochemical test .................................................................................................................. 42

2.4 Discussion ....................................................................................................................................... 49

2.5 Conclusion ...................................................................................................................................... 51

CHAPTER 3 Electrochemical Stability of Aluminum Current Collector in Aqueous

Rechargeable Lithium-ion Batteries ........................................................................... 53

3.1 Introduction .................................................................................................................................... 53

3.2 Experimental .................................................................................................................................. 55

3.2.1 Electrolyte preparation .............................................................................................................. 55

3.2.2 Electrochemical measurements ................................................................................................. 56

3.2.3 Inductively coupled plasma ....................................................................................................... 57

3.2.4 Surface Characterization ........................................................................................................... 57

3.3 Results and Analysis ...................................................................................................................... 57

3.3.1 Electrolyte stability window ...................................................................................................... 57

3.3.2 Cyclic Voltammetry ................................................................................................................... 59

3.3.3 Pitting potential measured by LSV ........................................................................................... 64

3.3.4 Chronoamperometry .................................................................................................................. 66

3.3.5 Surface morphology .................................................................................................................. 67

3.3.6 Raman Spectroscopy ................................................................................................................. 71

3.3.7 Concentration of dissolved Al3+ after CV ................................................................................ 75

3.4 Discussion ....................................................................................................................................... 76

3.4.1 The inhibiting effect of anions .................................................................................................. 76

3.4.2 Material selection for current collector in ARLB ...................................................................... 80

3.5 Conclusion ...................................................................................................................................... 81

CHAPTER 4 Effects of Sulfate and Nitrate Anions on Aluminum Corrosion in Slightly

Alkaline Solution ........................................................................................................... 83

4.1 Introduction .................................................................................................................................... 83

x

4.2 Experimental .................................................................................................................................. 85

4.2.1 ARLB electrolyte ....................................................................................................................... 85

4.2.2 Electrochemical tests ................................................................................................................. 85

4.2.3 Surface characterization ............................................................................................................ 86

4.3 Results and Analysis ...................................................................................................................... 87

4.3.1 Stability window of solutions measured by LSV ...................................................................... 87

4.3.2 Chronoamperometry .................................................................................................................. 88

4.3.3 Surface morphology of electrodes after chromoamperometry .................................................. 95

4.3.4 Effect of sulfate and nitrate anions on composition of surface passive film ............................. 96

4.4 Discussion ..................................................................................................................................... 105

4.4.1 The role of anion adsorption on corrosion kinetics ................................................................. 105

4.4.2 The influence of solution structure on corrosion kinetics ....................................................... 109

4.5 Conclusion .................................................................................................................................... 110

CHAPTER 5 Effects of Sulfate and Nitrate on the Electrochemical Behavior of Al-Fe and

Al-Fe-Si Intermetallic in Alkaline Solutions.............................................................. 112

5.1 Introduction .................................................................................................................................. 112

5.2 Experimental ................................................................................................................................ 115

5.2.1 Synthesis of Al-Fe and Al-Fe-Si alloy ..................................................................................... 115

5.2.2 Surface treatment of AA1100 .................................................................................................. 117

5.2.3 Electrolyte preparation ............................................................................................................ 117

5.2.4 Electrochemical testing ........................................................................................................... 118

5.2.5 Surface characterization .......................................................................................................... 118

5.3 Results and Discussion ................................................................................................................. 119

5.3.1 Intermetallic identification ...................................................................................................... 119

5.3.2 Potentiodynamic polarization .................................................................................................. 122

5.3.3 Surface morphology after potentiodynamic polarization ........................................................ 128

5.3.4 Raman Spectroscopy ............................................................................................................... 133

5.4 Discussion ..................................................................................................................................... 137

5.4.1 The role of sulfate and nitrate on anodic behavior of intermetallic alloys .............................. 137

xi

5.4.2 The role of Fe and Si on the electrochemical behavior of intermetallic alloys ....................... 139

5.5 Conclusion .................................................................................................................................... 140

CHAPTER 6 Summary ............................................................................................................ 142

6.1 Effect of Aqueous-based Cathode Slurry pH and Immersion Time on Corrosion of Aluminum

Current Collector in Lithium-ion Batteries .............................................................................. 143

6.2 Electrochemical Stability of Aluminum Current Collector in Aqueous Rechargeable

Lithium-ion Batteries .................................................................................................................. 143

6.3 Effects of Sulfate and Nitrate Anions on Aluminum Corrosion in Slightly Alkaline Solution

....................................................................................................................................................... 144

6.4 Effects of Sulfate and Nitrate on the Electrochemical Behavior of Al-Fe and Al-Fe-Si

Intermetallic in Alkaline Solutions ............................................................................................. 145

6.5 Novelty of this work ..................................................................................................................... 146

6.6 Limitations of this work............................................................................................................... 148

6.7 Future work .................................................................................................................................. 148

REFERENCES .......................................................................................................................... 150

CURRICULUM VITAE ........................................................................................................... 158

xii

LIST OF FIGURES

Figure 1.1 Calculated stability window of water with respect to pH values. ............................... 13

Figure 1.2 The intercalation potential of some electrode materials that could possibly be used for

aqueous lithium-ion batteries. Left: theoretical O2/H2 evolution potential versus NHE for

different pH in 1M Li2SO4 aqueous solution. Right: lithium-ion intercalation potential of

various electrode materials versus NHE and Li/Li+.25 ........................................................ 13

Figure 1.3 Elements that have been investigated as candidate materials for cathode current

collector in lithium based cells. ........................................................................................... 17

Figure 2.1 High-resolution SEM image of aluminum surface before testing. .............................. 35

Figure 2.2 EDS analysis shows the elemental composition of (a) intermetallic A, Al-Fe phase; (b)

intermetallic B, Al-Fe-Si phase. .......................................................................................... 36

Figure 2.3 High-resolution SEM images of aluminum surface after immersion tests, (a) 30

seconds at pH=12.98, (b) 1000 seconds at pH=11.56, (c) 1000 seconds at pH=12.98, (d)

10,000 seconds at pH=10, (e) 10,000 seconds at pH=11.56 ............................................... 38

Figure 2.4 Scan survey of aluminum foil immersed in NMC slurry for 10,000 seconds. ............ 40

Figure 2.5 Fitted XPS peaks of (a) bare aluminum and aluminum foils immersed in NMC slurry

for (b) 300 seconds, (c) 10,000 seconds. ............................................................................. 41

Figure 2.6 Fitted XPS peaks of aluminum foils immersed in sodium hydroxide solutions with (a)

pH=10, (b) pH=11.56, (c) pH=12.98 after 1,000 seconds and (d) pH=10, (e) pH=11.56

after 10,000 seconds. ........................................................................................................... 43

Figure 2.7 OCP vs time for 1085 aluminum in sodium hydroxide solution at pH 10, 11.56 and

12.98. ................................................................................................................................... 44

Figure 2.8 Potentiodynamic polarization curves of aluminum electrode under simulated sodium

hydroxide solutions at pH=10, 11.56 and 12.98. ................................................................. 45

Figure 2.9 Nyquist plots for aluminum 1085 immersed in slurries of (a) pH=10, (b) pH=11.56, (c)

pH=12.98 and relevant equivalent circuit model used to simulate the EIS spectra (d)....... 47

Figure 3.1 Linear sweep voltammetry measured on Pt foil in 2 M Li2SO4 electrolyte at pH 7. .. 59

Figure 3.2 Hydrogen and oxygen gassing potentials of (a) 2 M Li2SO4 and (b) 5 M LiNO3

electrolytes measured on Pt foil by LSV compared with theoretical values. ...................... 60

xiii

Figure 3.3 Cyclic voltammetry curves measured on Al foil in 2 M Li2SO4 electrolytes for four

consecutive cycles. .............................................................................................................. 62

Figure 3.4 Cyclic voltammetry curves measured on Al foil in 5 M LiNO3 electrolytes for four

consecutive cycles. .............................................................................................................. 63

Figure 3.5 Linear sweep voltammetry measured on Al foil in (a) 2 M Li2SO4 (b) 5 M LiNO3

electrolytes at different pH values. ...................................................................................... 65

Figure 3.6 Chronoamperometry measured on Al foil in (a) 2 M Li2SO4 and (b) 5 M LiNO3 at

different pH values for a period of 24 hours. ...................................................................... 68

Figure 3.7 Surface morphology of Al electrodes after cyclic voltammetry in 2 M Li2SO4 at (a)

pH 7, × 1000, (b) pH 7, × 50,000, (c) pH 11, × 1000 and (d) pH 11, × 50,000. ................. 69

Figure 3.8 Surface morphology of Al electrodes after cyclic voltammetry in 5 M LiNO3 at (a) pH

9, ×1000, (b) pH 9, ×50,000, (c) pH 11, ×1000, (d) pH 11, ×50,000, (e) surface corrosion

products at pH 11, ×2000 and (f) surface corrosion products at pH 11, ×5000. ................. 70

Figure 3.9 Morphology of pitting formed on Al electrode after chronoamperometry test in 2 M

Li2SO4 electrolyte at pH 11, (a) × 1000, (b) × 20,000. ........................................................ 71

Figure 3.10 Raman spectra of Al electrodes after cyclic voltammetry in (a) 2 M Li2SO4 and (b) 5

M LiNO3 at different pH values. The baseline samples shown are cleaned as-received foils

not subjected to any electrochemical exposure. .................................................................. 73

Figure 4.1 Chronoamperometry measured on AA1085 foil at 0.85 V in (a) Li2SO4 electrolytes

and (b) LiNO3 electrolytes at pH 11 for 24 hours. .............................................................. 89

Figure 4.2 Fitted linear portion of (a) Jpit vs t0.5 and (b) Jgr vs log(time) extracted from

chromoamperometry obtained in Li2SO4 and LiNO3 electrolytes. ..................................... 92

Figure 4.3 (a) Chromoamperometry measured on AA1085 foil in 0.1 M Li2SO4 at pH 11 and (b)

fitted linear portion of Jpit vs t0.5 under different anodic potentials. ................................. 94

Figure 4.4 (a) Chromoamperometry measured on AA1085 foil in 5 M LiNO3 at pH 11 and (b)

fitted linear portion of Jgr vs log(time) under different anodic potentials. ......................... 94

Figure 4.5 Surface morphology of AA1085 electrodes after chronoamperometry in 0.1 M Li2SO4

at 0.85 V for 24 hours (a) ×2000 and (b) ×10,000. ............................................................. 96

Figure 4.6 (a) Surface morphology and (b) cross sectional profile of AA1085 electrode after

chronoamperometry at 0.85 V for 24 hours in 5 M LiNO3 electrolyte. .............................. 98

xiv

Figure 4.7 XPS survey scan results of AA1085 electrode surface without Ar+ sputtering after

2000-seconds of chronoamperometry at 0.85 V. ................................................................. 99

Figure 4.8 High resolution XPS scans of Al 2p core-level peaks of AA1085 electrode after 0, 5,

10, 15, 20 and 25 minutes of Ar+ etching. ........................................................................ 101

Figure 4.9 High resolution XPS scans of S 2p peaks for AA1085 electrode after

chromoamperometry in 2 M Li2SO4 electrolyte at pH 11. ................................................ 102

Figure 4.10 High resolution XPS scans of Al 2p core-level peaks measured on AA1085 electrode

in 5 M LiNO3 of pH 11 after 0, 5, 10, 15, 20 and 25 min of Ar+ sputtering. .................... 104

Figure 4.11 The Raman spectrum obtained on deionized water. ................................................ 104

Figure 4.12 Comparison of Raman spectra obtained on (a) Li2SO4 electrolytes and (b) LiNO3

electrolytes with that of de-ionized water and LiOH at pH 11 solution. ........................... 105

Figure 5.1 BSE image of and the EDS spectra of the synthesized (a) AlxFey and (b) AlxFeySiz. 120

Figure 5.2 Secondary electron microcopy on the surface morphology of (a) AA1100 and (b)

surface treated AA1100. .................................................................................................... 120

Figure 5.3 XRD patterns of (a) AA1100 and (b) surface treated AA1100 with -Al (c)

synthesized Al-Fe with -Al2Fe and (d) Al-Fe-Si with -Al2FeSi, -Al4Fe1.7Si and

O-Al8Fe2Si ......................................................................................................................... 121

Figure 5.4 The potentiodynamic polarization curves of specimens tested in 0.001M LiOH, with

the addition of (a) 0.1 M and (b) 2 M Li2SO4. .................................................................. 124

Figure 5.5 The potentiodynamic polarization curves of specimens tested in 0.001M LiOH, with

the addition of (a) 0.1 M and (b) 2 M LiNO3. ................................................................... 125

Figure 5.6 The potentiodynamic polarization curves of the specimens in 1 M LiOH electrolyte.

........................................................................................................................................... 126

Figure 5.7 The potentiodynamic polarization curves of specimens tested in 1 M LiOH, with the

addition of (a) 0.1 M and (b) 2 M Li2SO4. ........................................................................ 127

Figure 5.8 The potentiodynamic polarization curves of specimens tested in 1 M LiOH, with the

addition of (a) 0.1 M and (b) 2 M LiNO3. ......................................................................... 128

Figure 5.9 The surface morphology of AA1100 electrode after polarization in 0.001 M LiOH, 2

M Li2SO4 electrolyte and the EDS mapping results. ......................................................... 130

Figure 5.10 The surface morphology and the EDS mapping results on surface-treated AA1100 in

xv

2M Li2SO4, 0.001 M LiOH. .............................................................................................. 131

Figure 5.11 (a) The surface morphology of Al2Fe after polarization in 1 M LiOH and (b) EDS

mapping results. ................................................................................................................. 132

Figure 5.12 The surface morphology and EDS mapping results of surface treated AA1100 in 1 M

LiOH, with the addition of 2 M LiNO3 and EDS mapping results. .................................. 133

Figure 5.13 Surface morphology and EDS mapping results of Al2Fe in 1 M LiOH, with the

addition of 2 M LiNO3.. .................................................................................................... 134

Figure 5.14 The Raman spectra of (a) AA1100, (b) surface treated AA1100, (c) Al2Fe and (d)

Al2FeSi0.67 in 1 M LiOH, 1 M LiOH with addition of 2 M Li2SO4, 1 M LiOH with addition

of 2 M LiNO3..................................................................................................................... 135

xvi

LIST OF TABLES

Table 1.1 Comparison of conductivity-weight-cost of possible materials used for current

collector in lithium-based battery. ....................................................................................... 18

Table 1.2 Cycling parameters for tested cells at 25 ℃ and 45 ℃. ............................................... 24

Table 2.1 Composition of AA1085 high purity Aluminum applied as cathode current collectors in

lithium ion batteries. ............................................................................................................ 34

Table 2.2 Surface aluminum/oxygen elemental ratio calculated from XPS peaks. ...................... 44

Table 2.3 Parameters obtained by fitting equivalent circuit model of 1085 aluminum immersed in

slurries of three pH values for EIS tests. ............................................................................. 48

Table 3.1 Hydrogen and oxygen gassing potentials of electrolytes measured by LSV on Pt foil. 60

Table 3.2 Average OCP and standard deviation for aluminum in test electrolytes. ...................... 61

Table 3.3 Band component analysis of Raman spectra obtained on Al electrode after cyclic

voltammetry in ARLB electrolytes. ..................................................................................... 74

Table 3.4 Concentration of Al3+ in electrolytes after cyclic voltammetry measured by ICP. ...... 76

Table 4.1 Gas evolution potentials and stability window of aqueous electrolytes measured by

LSV. ..................................................................................................................................... 88

Table 4.2 Kinetics parameters of the pitting corrosion on AA1085 in Li2SO4 electrolytes. ......... 95

Table 4.3 Fitted parameters for the repassivation rate on AA1085 in LiNO3 electrolytes. .......... 95

Table 4.4 The fitted band positions, individual peak areas and the band areas of deionized water,

LiOH solution and the aqueous electrolytes. ..................................................................... 103

Table 5.1 Chemical composition of AA1100 sheet. .....................................................................116

Table 5.2 The characteristic potentials obtained from potentiodynamic polarization curves. .... 122

Table 5.3 Raman bands measured on AA1100 and intermetallic alloy surface after

potentiodynamic polarization. ........................................................................................... 136

xvii

LIST OF NOMENCLATURE

R Resistance

L Inductance

Q Capacitance

Ceff Effective capacitance

R Resistance

Frequency

𝐸H+/H20

Standard hydrogen potential

𝐸O2/H2O0

Standard oxygen potential

𝐸H+/H2 Theoretical hydrogen evolution potential

𝐸O2/H2O Theoretical oxygen evolution potential

EO Oxygen evolution potential

EH Hydrogen evolution potential

Jc Current corresponds to the charging of the electric double layer

Jgr Current corresponds to the growth of passive film layer on aluminum due

to repassivation

Jdiss Current corresponds to the dissolution of passive film or electro-dissolution

of metal matrix

Jpit Current corresponds to the rapid dissolution of aluminum by pitting

t Time

A Rate of pit growth coefficient

xviii

ACKNOWLEDGEMENTS

I would like to express my deep appreciation and gratitude to my advisor, Professor Ben

Church for his patient guidance, support and mentorship throughout the course of my studies at

UW-Milwaukee. His profound knowledge and encouragement has made my PhD study a

learning and thoughtful and experience. I’m truly fortunate to work on research together with

him.

I would also like to thank my committee members, Professor Pradeep Rohatgi, Professor

Nidal Abu-Zahra, Professor Junhong Chen and Professor Ilya Avdeev for serving on my doctorial

committee. Their kind guidance, thought-provoking suggestions have helped me complete my

dissertation successfully.

I’d like to give special thanks to Professor Deyang Qu and his post-doc Dong Zheng from

UWM Mechanical department for their assistance in ICP set ups and analysis. I’m grateful to Mr.

Steve Hardcastle from UWM AAF for his assistance in XPS analysis and Professor H.A. Owen

from UWM Biology department for her guidance on SEM and BSE imaging. Their valuable input

has led this more research insights into this work.

I am grateful to my dear colleagues and friends for sharing the time at UW-Milwaukee.

Particularly, I’d like to thank Yangping Sheng, Aruna, P Wanninayake, Yihan Xu and Robert Lee

Hanson for their encouragement and assistance in work.

At last, I would like to express my greatest appreciation to my wife and my parents, for

their unconditional love and support to me in this wonderful journey.

1

CHAPTER 1 Introduction

1.1 Background and motivation

Aluminum is the most abundant metal element in earth and it is also the second most

consumed metal in the world. Aluminum and its alloys are well known for lightweight, high

reflectivity, high electrical and thermal conductivity.1 A thin layer of oxide passive film is

naturally formed on aluminum surface, protecting aluminum from corrosive media attack. These

unique properties make it a perfect material for both conventional and novel applications.

Depending on the amount of impurities, aluminum is classified into extreme high purity

aluminum and commercial purity aluminum (primary aluminum). The aluminum purity level

affects many of its properties especially upon electrical resistivity and thermal conductivity.1 In

lithium ion batteries (LIB), commercial grade aluminum is extensively used as the current

collector for lithium oxide cathode electrode. Compared to stainless steel, nickel mesh etc., it has

many advantages, such as relatively low cost, high electronic conductivity and availability as

high purity thin foils or plates.2 The current collectors has to be electrochemically stable in

contact with the cell components over the operating potential window of the battery. In most

electrolytes, aluminum is also stable up to 4.5V vs. Li+/Li. Aluminum may also be coated on

insulating substrates by physical deposition, which allows much less use of metal. Thus

commercial aluminum is commonly considered as the material of choice for high voltage (>3.5

V) lithium ion batteries.

In lithium ion batteries, current collector is not involved in the lithium

intercalation/de-intercalation reactions so it is considered as an inactive component, which is

among the many factors influencing the cell gravimetric energy densities. Recently, the studies

2

carried out on current collectors are mostly focused on its degradation in contact with various

chemistries of LIB components. The standard redox potential for Al/Al3+ is -1.676V (SHE) in

acidic solutions.2 Although aluminum is expected to be thermodynamically stable due to a thin

layer of oxide passive film, it may subjected to corrosion during the continuous contact with

complex chemistries in LIB systems. The degradation of Al current collector may result in a

series of problems, including adversely increase the electrical resistance, generate corrosion

products that contaminate active materials, leading to the attenuation of the battery’s electrical

performance, life or even safety.

The development of renewable energy requires new energy storage systems with high

energy density, high cycling rate, high round-trip efficiency, long service life, enhanced safety

performance and reduced manufacturing cost. To meet these requirements, the materials

selection and materials design is important during the development of the new technologies.

Aluminum, as the most abundant metallic material, with lightweight, low cost, high conductivity,

corrosion resistance and considerable mechanical strength, has been widely adopted as

conductive substrate material in energy storage systems. Any sudden failure or long-term

degradation of the aluminum current collector is big concern because it’ll adversely affect the

electrical performance, capacity, life, and safety. The factors that might bring such issues include:

(i) the electrical resistance increases to a point that the continuity is lost. (ii) the active electrode

materials are attacked by the corrosion products. (iii) introduction of contaminants that will react

with active materials.3 Understanding the electrochemical stability, corrosion mechanisms and

corrosion kinetics of current collectors in the service life environment it’s exposed to, helps

ensure the proper use and selection of aluminum and its alloys, avoid possible catastrophic

3

failure caused by the corrosion in cost-effective lithium ion batteries. This study is also rendered

very necessary by the vast use of aluminum and its alloys in aerospace, automotive and structural

applications. The objective of this work was to understand how the continuity of aluminum

current collector is affected by aqueous based cathode slurry and ARLB electrolytes, and if the

reliability and the service life of batteries would be compromised. The corrosion behavior and

explain the corrosion mechanism of current collectors exposing to aqueous slurries and

electrolytes was investigated. As it is very difficult to conduct experiment process in functional

cells, simulated electrochemical conditions were applied without considering more complex

situations with the active electrode materials involved. Excess electrolyte is used to minimize the

influence from effects from concentrated corrosion product but notably it might exaggerate the

effects of aqueous solutions.

1.2 The role of structural and compositional features on Al corrosion

1.2.1 The protective surface passive film

Although aluminum is one of the most active metals (-1.67V vs. SHE), the oxidation rate

is extremely low at room temperature or even up to 600 ℃.4 This is due to the fact that a layer of

oxide passive film is naturally formed on aluminum surface under ambient conditions, which is

so called passivation phenomenon. Passivation plays important role in various technological

applications, such as catalysts, sensors, lubrication, dielectrics and corrosion protection. From a

corrosion point of view, passive films should be stable or exhibit very low rate of dissolution in

the passive potential range. The break down potential for the passive films should be as high as

possible. The oxide layer formed on aluminum is non-uniform, very thin, only a few nanometers

thick.5 It is crucial for the corrosion resistance property of aluminum. On aluminum alloys, this

4

protective film is, however, very susceptible to pitting corrosion due to the existence of

intermetallic particles, which leads to accelerated corrosion of underlying aluminum matrix.

J. D. Baran et al. described that the reason for the limited thickness of aluminum passive

film is due to the decreased oxygen absorption energy, which prohibit the supply of oxygen

molecules.4 The oxidation of aluminum at low temperature relies on an electrochemical

mechanism as opposed to thermal activation for high temperature oxidation. Aluminum

oxidation starts with the dissociative chemisorption of O2 from gas phase and charge transfer

from metal to the oxygen. The subsequent passive film growth involves the absorption and

dissociation of oxygen on bare aluminum metal surface. Upon the oxide film is covered on

aluminum surface, cations and anions as well as electrons transport through the growing oxide

film. The ionic diffusion through the oxide film is controlled by the electric field established by

tunneling electrons due to the potential difference across the passive film (Mott potential). As the

rate of electron transport through the oxide film decreases exponentially with the film thickness.

The charge neutrality of coupled currents of electrons and cations means that the thickness of

oxide film is limited at low temperatures.

The aluminum oxide layer formed at low temperature is amorphous alumina film. Bulk

Al2O3 is an insulator with a band gap of 8-9 eV. The passive film on aluminum exhibits a band

gap of 3 eV. During the oxidation process oxygen anions are close packed with the aluminum

cations over the octahedral and tetrahedral interstices. The passive film shows a deficiency of Al

cations thus it is considered that the oxide-film growth is limited by the cation migration under

the influence of Mott potential VM. The oxide film growth rate is described by:

dL

dt= Ω n ν exp(

−U + qaVM/L

kT)

5

Where L represents the film thickness at time t, Ω is the volume of oxide film formed per

transported cation, n is the number of cations per unit area that jump through the diffusion barrier

U, q is the charge of migrating ions, a is the distance between two adjacent potential minima, 𝜈

is the attempt frequency for ion migration, k is the Boltzmann constant and T is the temperature.

The effective limiting thickness of the oxide film Llim is defined as when𝑑𝐿

𝑑𝑡≤10-5Å/s, which

means the oxide growth rate is less than one oxide monolayer per 105 s. The limiting oxide layer

thickness Llim is given by,

1

𝐿𝑙𝑖𝑚= −

𝑘𝑙𝑛(10−15/Ω 𝑛 𝜈)𝑞𝑎𝑉𝑀

𝑇 −𝑈

𝑞𝑎𝑉𝑀

Generally, the Mott potential VM, is determined by the potential difference between the

metal-oxide work function Φ𝑚 and the oxygen-oxide work function Φ𝑜. VM = (Fm − FO)/e,

where e is the electron charge.4 In a recent work of Na Cai et al., it is described that the mobility

of Al cations is affected by the oxygen pressure, thus influence the overall oxidation kinetics.6

Depending on various environmental chemistries and conditions, the passive film formed on

aluminum may consists of oxides, oxy-hydroxides, hydroxides.7 The role of aluminum

intermetallic particles

As the major impurities in pure aluminum, Fe and Si are usually dissolved in aluminum

matrix or forms intermetallic phases. These particles are formed during solidification and are not

dissolved in the following thermal-mechanical processing.8 Intermetallic in aluminum are either

natural impurities or intentionally developed to achieve desired mechanical properties. Because

some particles do not play a pivotal role in the mechanical properties of aluminum alloys, the

precise mechanical properties of these particles are still not well known. However, it was

described that the existence of these impurities results in high hardening rate in aluminum.

6

1xxx alloys contain Al6Fe and Al3Fe as natural impurities. Some intermetallic adversely

affects pitting resistance, such as intermetallic with Cu and Fe in 2xxx and 1xxx alloys. The

influence of the intermetallic primarily depends on the potential difference between the particle

and the matrix metal in a solution. Intermetallic that is more electrochemically stable than matrix

act as cathode and the matrix metal undergoes anodic dissolution.9 In high-purity aluminum,

Al-Fe and Al-Fe-Si are identified as the primary intermetallic particles. There are three

equilibrium phases in Al-Fe-Si ternary phase diagram, they are Al3Fe, α and β phase.10 But over

then different metastable phases could be formed under actual solidification conditions. The

composition, size, distribution and volume fraction of intermetallic phase will affect the

mechanical properties of foils significantly. β phase transforms into α phase during intermediate

annealing, resulting in a favorable decrease of particle size. The precipitation formed during

annealing also contributes to an impurity concentration change in aluminum matrix, which at the

same time results in matrix lattice parameter change.

The alloying elements in intermetallic particles make them electrochemically different

from the surrounding phases. The electrochemical response and activity of most metals and

alloys are remarkably different with the change of solution pH due to the existence of

intermetallic particles.11 It results in ramifications in the corrosion morphology on aluminum

alloys with various types and composition of intermetallic particles, which is known as galvanic

corrosion. Galvanic corrosion is ranked as the leading cause of all corrosion experienced by

aluminum in electrical applications. It occurs when two metals come to direct contact with one

another while immersed in an electrolyte. The difference in electromotive force between metal

7

and its intermetallic phase results in current flow from the cathode to the anode, which lead to

the oxidation of anode.

In near neutral solutions, the pitting corrosion is influenced by the intermetallic particles.

There are mainly two types of corrosion that were detected, one is called “circumferential”,

appear as a pit around more or less inert particle, with the corrosion happened mostly in the

matrix (trenching).12 Another type of pit is grown deep in matrix and may contain some

intermetallic particles remnants inside, which is caused by the preferable dissolution of the

intermetallic particle. Intermetallic affects the homogeneity of passive film on aluminum and

serve as cathodic sites for pit nucleation. The intermetallic particles dissolve selectively and the

remnants after the particle dissolution such as Fe and Cu, are even more cathodic than the

intermetallic. At a potential below the pitting potential, the deterioration of passive film

properties results in the rupture of passive film and produces metastable pits. During pit growth,

the interface events influence growth process, while the physical and chemical properties of

passive film influence the initiation of pits but play a secondary role in pit growth. Noble

intermetallic particles act as cathodic sites even when the electrolyte is deaerated, reduction of

residual oxygen and a low level of hydrogen evolution occur and cause anodic dissolution of Al

adjacent to the intermetallic particles. The magnitude of the corrosion potential difference

between the intermetallic and aluminum matrix can be used to estimate the corrosion behavior of

aluminum alloys. But the corrosion potential is not adequate for the understanding of corrosion

mechanisms. Additional information on the electrochemical behavior, the structure of the alloy is

required.

8

The influence of the Al3Fe intermetallic on aluminum pitting was studied in prior works.

Rajan et.al reported that the anodic and cathodic reactivity of Al-Fe alloys are both dominated by

the distribution of Al3Fe intermetallic particles, even the Fe is at a very low compositional level

of 0.04%.13 The greater the number of intermetallic particles, the higher the cathodic reactivity.

The cathodic reactivity increases with the iron content of the alloy. Nisancioglu reported that at a

potential close to the corrosion potential, aluminum in Al3Fe preferably dissolves and the surface

of Al3Fe became rich in iron.14 A transient behavior for the corrosion potential of Al3Fe during

the first 30-200 min of immersion was found in 0.1 M NaOH. A protective layer of Fe3O4

formed on the intermetallic. The iron rich layers are highly porous, which can act as catalytic

sites for oxygen reduction. The presence of Mn or Si in the phase reduce the effect of iron on

both anodic and cathodic reaction rates. In general, the results show that Al3Fe and Al-Fe-Si

presents an increased cathode activity due to the selective dissolution of Al. Hassan et. al

reported that Al, Al6061 and Al-Cu alloy presented different corrosion resistance in alkaline

solutions, which is associated to the effects of alloying elements. The high percentage of Cu

(4.8%) in the Al-Cu alloy decreased the corrosion resistance of the alloy. In Al6061, the presence

of Mg and Si leads to the formation of Mg2Si phase, which has no pronounced influence on

electrode potential.15 The general effect of Mn and Si in intermetallic is to reduce both the anodic

and cathodic currents significantly due to passivating effect. The addition of Si reduces the

anodic oxidation peaks, reduces the rate of hydrogen evolution, and shifts the corrosion potential

to more negative potentials. 14

9

1.3 Aqueous based lithium-ion battery system

Rechargeable lithium ion batteries have been widely used as commercial energy storage

systems for portable equipment, electronic devices and higher power applications (e.g. electronic

vehicles). Due to its advantages in superior performance, flexibility in design and high energy

density etc., LIB has been considered as the best option for energy storage system used in

electric vehicle (EV) or hybrid electric vehicle (HEV).16 Beck and Ruetschi proposed the “three

E” criteria as the requirements for good energy storage systems, which highlights energy,

economics and environment. Specifically, the energy storage system should have high energy

density (high energy content with respect to unit weight and volume), economic (with low

fabrication costs and long cycling life), environment (safe to be used, nontoxic and high

reliability).17

The reason that lithium ion batteries could work is because of the ability of Li ions to be

inserted or extracted from positive materials. Lithium atoms are inserted into a host solid as guest

atom with only slight and reversible changes in the host material. The host materials are usually

layered material like graphite or tunnel structure compounds e.g. LiMnO4 and LiFePO4. The

intercalation is happened because of the lowered chemical potential of lithium during its

insertion into the host material. When the cell is discharged, the intercalated lithium dissociates

into ions and electrons move to the positive electrode through the electrolyte and the electrical

circuit. The ions and electrons meet at the surface of host material and they will be intercalated

into the material. The cell voltage can be calculated by the difference between the potential of Li

in intercalation hosts divided by the charge.18

1.3.1 Advantages of aqueous based lithium ion battery system

10

During the fabrication of lithium-ion battery positive electrodes, solid active materials are

blended with binders, solvent, and conductive carbon. The slurry paste is then coated on

aluminum foil, dried and compressed to generate a cathode coating of controlled loading of

active materials. The commercial lithium-ion batteries employ an organic solvent,

n-methylpyrrolidone (NMP) to prepare the slurry. But for the concerns of processing

requirements, production cost and environmental issues, manufacturers may move away from

organic NMP solvents and instead utilize aqueous based slurries. The most commonly used

aqueous binder is the aqueous emulsion of styrene butadiene rubber (SBR) blended with

water-soluble sodium carboxymethyl cellulose (CMC) 19. The use of aqueous slurry brings

distinct advantages for lithium ion battery manufacturing, such as elimination of toxic, volatile

organic binder, and lower production cost.

Organic electrolyte is extensively used in commercialized lithium ion batteries. In

commercial lithium ion batteries, the most commonly used electrolyte is derived from solutions

of lithium salt in non-aqueous solvents, such as alkyl carbonates or solvent blend.20 However, the

organic based electrolyte system has a lot of drawbacks. First of all, the organics are highly toxic

and flammable, which may cause safety hazards if the battery is overcharged or self-circuited.

Besides, the ionic conductivity of the organic electrolyte is poor, which is two orders of

magnitude lower compared to aqueous electrolyte. Generally, the conductivity of organic

electrolytes dissolving LiPF6 is 20 mS/cm at room temperature, however the conductivity of

aqueous based electrolytes are close to 1 S/cm. Due to the limited conductivity, the electrode in

organic based lithium ion battery must be thin. More importantly, the fabrication cost of organic

11

based LIB is high. These drawbacks have limited its application in large-scale energy storage

systems.21

In such scenario, the possibility of producing lithium ion battery with aqueous based

system has been considered. In 1994, Dahn’s group first introduced the concept in which VO2 is

used as negative electrode and LiMn2O2 as positive electrode in 5M LiNO3 aqueous

electrolytes.18 Aqueous electrolytes own several advantages over organic electrolytes. They are

non-flammable thus offers much better stability and safer performance. Electrolyte of water

solution and the separator sheets used are both much cheaper compared to organic electrolytes.

In addition, the conductivity of aqueous electrolytes is significantly higher, which brings higher

rates and lower voltage drops due to electrolyte impedance. It has attracted wide attention

because its good cycling performance and super-fast charge performance, which can be

comparable with filling gasoline for engine cars. One challenge in aqueous based lithium ion

battery technology is the H2/O2 evolution reactions in aqueous electrolyte. It is known that

capacity of the electrode material should be used as much as possible before electrolyte

decomposition. But the evolution of H2/O2 inevitably happens at full charge stage.22 It results in

pH change nearby the electrode and affects the stability of the electrode materials. In organic

electrolyte systems, although it was reported that the decomposition of electrolyte occurs at high

voltage, a protective film (SEI layer) is formed between the active material and the electrolyte

and reduce the further decomposition. But there are no such protecting mechanisms in aqueous

based LIBs.23 Due to the limited operating potential range within the electrochemical window of

water, the main disadvantage of aqueous lithium ion battery is the low energy density compared

to the conventional lithium ion battery.

12

1.3.2 Stability window of aqueous electrolytes aqueous based lithium ion battery

Theoretically, an aqueous lithium-ion battery can be assembled by combining a lower

potential lithium-accepting anode and a higher potential lithium-source cathode within the O2/H2

evolution potential range.

From basic thermal dynamics, the equilibrium of H+ and H2 in aqueous solution is

described as the following equation:

2𝐻+ + 2𝑒− ⇌ 𝐻2(𝑔) ↑

In high pH solution, the equilibrium relationship is as follows:

2𝐻2𝑂 + 2𝑒− ⇌ 𝐻2(𝑔) ↑ +2𝑂𝐻

−

Thus, for hydrogen evolution, the dissociation potential can be illustrated by Nernst equation:

𝐸𝐻+/𝐻2 = 𝐸𝐻+/𝐻20 +

𝑅𝑇

𝑛𝐹𝑙𝑛

[𝐻+]2

𝑃𝐻2

In equilibrium condition at 25℃, the equation becomes:

𝐸𝐻+/𝐻2 = 𝐸𝐻+/𝐻20 − 0.059pH

The decomposition of water into hydrogen is favored when the potential is lower than the

hydrogen evolution potential. But when potential becomes more positive or noble, water will

decompose into its other constitute oxygen, as illustrated in equations for the acid form of the

process,

𝑂2 + 4𝐻+ + 4𝑒− ⇌ 2𝐻2𝑂

A Nernst equation is used to describe the potential in standard conditions of temperature and

oxygen partial pressure of value unity.24

𝐸𝑂2/𝐻2𝑂 = 𝐸𝑂2/𝐻2𝑂0 +

𝑅𝑇

𝑛𝐹𝑙𝑛(𝑃𝑂2[𝐻

+]4)

13

𝐸𝑂2/𝐻2𝑂 = 𝐸𝑂2/𝐻2𝑂0 − 0.059pH

The potential range of the stability window is shown in figure 1.1. According to the operating

potential range, the possible candidate materials for cathode and anode electrodes in aqueous

based lithium ion battery are shown in the following graph (figure 1.2).

Figure 1.1 Calculated stability window of water with respect to pH values.

Figure 1.2 The intercalation potential of some electrode materials that could possibly be used for

aqueous lithium-ion batteries. Left: theoretical O2/H2 evolution potential versus NHE for

different pH in 1M Li2SO4 aqueous solution. Right: lithium-ion intercalation potential of various

electrode materials versus NHE and Li/Li+.25

14

Previous study shows that materials with voltage versus Li/Li+ higher than 3.3 V are

generally stable. The intercalated potential of lithium-ion is below 3.3V versus Li/Li+ when

being as negative or anode materials. As the aqueous based LIB operates in air, the intercalated

lithium may react with H2O and O2 in the following way,

𝐿𝑖(𝑖𝑛𝑡𝑒𝑟𝑐𝑎𝑙𝑎𝑡𝑒𝑑) +1

4𝑂2 +

1

2𝐻2𝑂 ⇔ 𝐿𝑖

+ + 𝑂𝐻−

The potential of a LIC, V(x) can be calculated with the equation: V(x) = −1

e(uLi

int(x) − uLi0 ),

Where 𝑢𝐿𝑖𝑖𝑛𝑡(𝑥) is the chemical potential of intercalated Li in cathode, 𝑢𝐿𝑖

0 is the chemical

potential of Li in Li metal.18

1.3.3 Recent progress on development of aqueous based lithium ion batteries

In 1999 Mohan Rao et al. reported for the first time that the lithium deficient Li1-xNiO2

has a chemically reversible electrochemical proton intercalation in alkaline aqueous electrolytes.

It is also reported that the electrochemical stability of LiCoO2 is dependent on the hydrogen ion

concentration a lot. The cathode material is stable when pH is less than 9 or when it is higher

than 11. The electrochemical performance of LiNi1/3Co1/3Mn1/3O2 in aqueous solution is also

similar to that of LiCoO2. Yuan et al. studied the electrochemical behavior of LiMnO4 electrodes

in 2M Li2SO4, 1M LiNO3, 5M LiNO3 and 9M LiNO3 aqueous electrolytes. The results show that

the LiMnO4 electrode in 5M LiNO3 electrolyte shows good electrochemical performance in

terms of specific capacity, rate ability and charge/discharge cyclability.

Recently Mentus et al. reported that the addition of vinylene carbonate into aqueous

LiNO3 solution effectively improved the cycle life of Li1.05Cr0.1Mn1.85O4 in aqueous electrolyte

solution. Nurhaswani alias described the intercalation of lithium ions for carbon coated LiFePO4

15

in an aqueous lithium ion battery. The electrochemical behavior of LiFePO4/C vs. air electrode

as counter electrode was also evaluated in a 5M LiNO3 solution.26

Riccardo Ruffo et al. reported that lithium insertion and extraction can occur in LiCoO2 in

LiNO3 aqueous solutions. With a concentration of 5 M LiNO3, fast kinetics and good cycling

behavior at high rates were found.27 In the above aqueous based lithium ion battery research,

instead of aluminum, nickel mesh, stainless steel mesh are applied as the current collectors. The

possible reason is due to the severe degradation of aluminum with the adding of aqueous based

electrolyte.

1.4 Corrosion of aluminum current collector in lithium ion battery

Corrosion is typically defined as the deterioration of metals. In a simple term, corrosion is

the inherent tendency of a metal to revert from processed, metallic state into more nature state.

For example, iron and steel tends to combine with other elements to return to their lowest energy

states. Generally, the corrosion process can be described as chemical reaction or electrochemical

reaction between metal and the contacting media, which leads to loss of material and its

properties. Corrosion results in wasting away of materials or sudden failure of metal components

so it has great impact on economy. In USA, the cost of corrosion on economy is in the vicinity of

3-4% of Gross National Product. These costs are possible to be reduced by application of broader

materials design and corrosion protection technologies. The primary methods for protecting

material from corrosion include careful material selection, coating, inhibitors, cathodic

protection and improvement on part designing.

1.4.1 Requirements for current collectors in lithium-ion battery

16

Current collector is not involved in the oxidation/reduction battery reaction thus it is

considered as inactive mass and volume in lithium ion batteries. A few properties are required

for material to be used as qualified current collector. First, to achieve high gravimetric

volumetric energy densities, current collectors that are thinner, lighter with compatible

mechanical strength are usually preferred. The shape and mechanical property requirements

might be costly depends on the material selected and the processing method. Secondly, within

the entire operating potential, current collector must be chemically and electrochemically stable.

Ideally current collector should not react with any other components in the battery. Third, it has

to be adhesive to cathode mix, including the cathode material, binder and conductive material.2

Evaluation on the stability of current collectors is usually carried out on bare current collectors in

direct contact with media using electrochemical methods. Various methods, such as EIS, cyclic

voltammetry and potential step measurements have been applied in these studies although there

is no standard upon the evaluation methods.

A. H. Whitehead et. al2 reviewed the materials that haven been studied and used for

cathode electrode in lithium based batteries. In the design of these battery devices, a current

collector is used to make current flow between electrodes. Most of the materials studied were

metallic, as listed in figure 1.3. Some metallic materials in the table are not suitable as current

collector due to low conductivity, high cost and instability. A rough comparison of the

weight-conductivity-cost of these materials is presented in table 1.1.

17

Figure 1.3 Elements that have been investigated as candidate materials for cathode current

collector in lithium based cells.

Until now aluminum is considered as the best candidate materials as current collectors for

cathode compared to other materials such as Ni, stainless steel, Ti etc. From thermodynamic

basics, aluminum will corrode during battery cycling because the standard electrode potential of

aluminum (1.39V vs Li/Li+) is lower than the operating potential of the cathode electrode. Due to

the formation of passive film, however, aluminum is kinetically stable in many conditions.

Aluminum could withstand without corrosion until high voltage. However, it could not be used

as anode current collector because of its reaction with lithium at potentials near the intercalation

potential.

Iwakura et al.28 compared the electrochemical stability of different metal foils (Al, Cu, Fe,

Ni SUS304 and Ti) in 1M solution of LiClO4 and EC/DMC (1:1) electrolytes using cyclic

voltammograms and EIS. The anodic current on Al and Ti foils are both very small over the

measured voltage range compared to Fe and Cu in the electrolytes. The anodic current on

aluminum current collector is also correlated with the impurity level and there is a proportional

relationship between the current and impurity level. The content of metal ion in the electrolytes

were examined after polarized at 4.5V vs Li/Li+ for 10h. Detectable amount of metal ions was

only found on Fe foil, which indicates the deterioration of Fe electrode. The study also showed

18

that the content of aluminum ion after the polarization test increases with higher impurity levels.

Aluminum of different purity level, from 96.9% to 99.9% was tested in LiClO4/EC/diethyl

carbonate electrolyte. Aluminum with Fe, Mg, and Mn as impurities showed higher

chronoamperometric current compared to aluminum with higher purity level. Thus it draws to

the conclusion that electrochemically aluminum with high purity is the most suitable material as

current collector for the positive electrode.

Table 1.1 Comparison of conductivity-weight-cost of possible materials used for current

collector in lithium-based battery.

Material Relative conductivity per unit

volume

Relative Conductivity per unit

mass

Relative

conductivity per

unit cost

Ag 1.05 0.9 0.01

Cu 1 1 1

Au 0.7 0.33 0.00008

Al 0.4 1.3 2

Mo 0.31 0.27 0.01

W 0.29 0.13 0.02

Zn 0.28 0.36 0.8

Ni 0.24 0.25 0.05

Fe 0.17 0.2 2

Pt 0.16 0.067 0.000008

Cr 0.13 0.16 0.05

Ta 0.13 0.072 0.001

304SS 0.1 0.1 0.1

316SS 0.1 0.1 0.07

Ti 0.04 0.079 0.02

SiC 0.012 0.032 0.001

Mn 0.009 0.01 0.01

C pyrolytic

graphite ~0.007 ~0.03 —

C graphite ~0.0003 ~0.0012 ~0.0005

C black ~0.00001 ~0.00004 ~0.00002

19

Corrosion of aluminum may occur when it is exposed to electrolytes or electrolyte

solvents in lithium ion batteries. Current collector is assumed well protected when cathode

material is coated on current collectors, but the cathode material is manufactured into us

structure intentionally to increase the cathode/electrolyte interfacial area so intercalation and

deintercalation of lithium ions can proceed during the discharge/charge cycling. The localized

corrosion of current collector is also ascribed to the through thickness porosity.29 Although the

air formed aluminum oxide passive layer is somewhat protective, it is not capable of protecting

aluminum against oxidation at high potentials. For instance, in aqueous solution of 1 M NaNO3,

KSCN, and 1 M NaCl, pitting corrosion of aluminum was found to start at 5.0V, 4.5V and 2.6V

respectively. The passive layer may not form or dissolved under some conditions. Aluminum is

found to dissolve above 3.7 V vs Li/Li+ in LiAlCl4/SO2. Passive film may also form and be

stable in anhydrous organic solvents, such as LiBF4 or LiClO4 in ethylene carbonate

(EC)/propylene carbonate (PC) with LiMn2O4 as the cathode.30

1.4.2 Factors that influence aluminum corrosion in non-aqueous based lithium-ion battery

High reliability and longer service life is required for advanced lithium ion batteries. For

long-term applications, the degradation of cell materials has been an issue because of its

possibility of adversely affecting the electrochemical performance, capacity, life and safety.

Corrosion of aluminum current collector in lithium ion batteries is considered to be a factor that

affects the long-term stability of lithium-ion batteries. The degradation of Al current collector,

especially localized corrosion, may greatly affect the calendar life and cycling performance of

the batteries.31 The corrosion of current collector cause many problems: (i) the corrosion

passivates the cathode active material, (ii) the non-soluble corrosion products increases the

20

electrical resistance, (iii) soluble corrosion products contaminate the electrolyte thus increase

self-discharge rate and (iv) Al3+ ions result from anodic dissolution migrate into the cell and

reductively deposit on anode.31

Generally, the corrosion of aluminum current collector may be caused by electrochemical

reactions between aluminum and other components in the battery, such as the electrolyte solvent,

lithium salt and the cathode active materials. A lot of studies have been done on corrosion of

aluminum current collector in organic based lithium ion batteries. The intrinsic corrosion

resistant properties of aluminum are usually evaluated in simulated electrochemical conditions

that are not encumbered by more complex conditions with the presence of active electrode

materials. The corrosion of aluminum was extensively evaluated in non-aqueous LIB system

with different combination of electrolyte and electrolyte solvents.

(i) Effect of electrolyte salt on corrosion of aluminum current collector

Most of the commercial lithium-ion battery use LiPF6 as electrolyte salt. Corrosion of

aluminum current collector due to LiPF6 salt was detected in many studies. One possible

mechanism for the aluminum corrosion in LiPF6 contained electrolytes is proposed to be crevice

corrosion ascribed to the cathode coating. The ratio of surface area of metal inside the crevice

and the volume of solution in the crevice is closely related to the severity of corrosion.32 The

inevitable existence of traceable amount of water in LiPF6 was also proposed to be the cause of

corrosion problems. PF6−

can react with trace water and generate HF, which react with the active

materials thus bring in more water and also continuously corrode aluminum current collector.

Krause et al reported very high corrosion rates of aluminum was found in PC electrolyte

containing LiCF3SO3 and LiN(CF3SO2)2 salts when aluminum is potentiostatically polarized at

21

+4.2V. H. Yang et al.33 further investigated the stability of aluminum current collector in

propylene carbonate (PC) solutions containing 1M of different lithium salts with electrochemical

quartz crystal microbalance (EQCM). The mass of aluminum electrode and the charge transfer

involved in the corrosion process was measured as a function of potential and time. The results

showed that aluminum corrosion occurred in PC containing LiN(CF3SO2)2, LiC(CF3SO2)3, and

LiCF3SO3. In LiPF6 or LiBF4 contained electrolytes, corrosion barely happens due to the

formation of protective film. Anodic polarization tests in EC/DME electrolyte showed that the

corrosion resistance of aluminum current collector in contact with different salts ranks in the

following order: LiCF3SO3

22

which suggests that P and B species deposition offers more stability for aluminum compared to F

species.2,34 Research was also carried out on evaluating the stability of aluminum passive film in

LiTFSI contained electrolytes. Due to the acid-base property of LiTFSI and the stereochemistry

of the anion, it lowers the stability of aluminum.35

(ii) Effect of cathode materials on corrosion of aluminum current collector

The extent of aluminum corrosion is influenced by the type of cathode active material in LIB.

For instance, compared to LiFePO4, LiNi1/3Co1/3Mn1/3O2, higher rates of corrosion occurred on

LiMnO2 cathodes.32 It was reported that severe corrosion was detected on aluminum foils in

LiMnO2 and LiFePO4 cathodes. Comparable less corrosion is found on LiFePO4 electrode

because the operating voltage is lower than that of LiMnO2. The reason still relies on the

corrosive electrolyte salts. LiPF6 is expected to decompose more at higher voltage thus higher

operating voltage will deteriorate aluminum during the long term battery cycling.

(iii) Effect of electrolyte solvent

J. W. Braithwaite et al.3 studied the corrosion of aluminum current collector in 1:1

propylene carbonate and diethylene carbonate (PC:DEC) electrolyte and 1:1 mixture of ethylene

carbonate and dimethyl carbonate (EC:DMC) with 1M LiPF6 solvent. After 40 cycles, both

general corrosion and scattered localized corrosion are found on aluminum surface in both

electrolytes. Higher rate of corrosion occurs in EC:DMC electrolyte condition. Under increased

charge potential condition, corrosion resistance of aluminum is found decreased. Possible factors

that influence the pitting behavior of aluminum was proposed, including cycling aging, charge

potential, alloy composition, water contamination and temperature. Aluminum current collector

23

was proved to have excellent corrosion resistant performance in 1M LiPF6 in 1:1 EC+DMC and

in other electrolyte systems, including EC+EMC, EC+EMC+PC.

1.4.3 Adverse effect of corrosion of current collector on lithium-ion battery performance

Deterioration of battery performance is a function of many factors, including operating

temperature, coating quality of active materials and electrolyte composition, etc. Corrosion of

aluminum current collector results in corrosion pitting, cracks in aluminum foil, contamination of

aluminum ions released into electrolyte (as high as 2700 ppm) and even possible mechanical

degradation of cathode. There is no doubt that the corrosion will lead to a continuous increase in

the internal resistance of the battery, with considerable loss of apparent capacity. Thus it is

expected corrosion will significantly affect the capacity fade and power fade in lithium ion

batteries. Many studies have been carried out to evaluate the performance of individual parts of

lithium ion batteries, however, few have been found on influence of aluminum current collector

corrosion.

Xueyuan Zhang et. al30 evaluated the corrosion of current collector in lab assembled cells

which use different types of cathode materials using LiPF6 as the salt. The cells are disassembled

after life cycling test and the corrosion on current collector surface were observed by optical

microscope. It is found that a small amount of corrosion may trigger significant capacity loss (as

high as 20%), which might due to the loss of contact between cathode material and current

collector. The charge/discharge performance is not directly proportional to the corrosion extent.

However, the cells with poor performance with severely corroded current collectors were cycled

only a few times and then failed (less than 200 charge/discharge cycles). Zhang et. al also found

that the corrosion of Al plays an important role in the self-discharge of Li/LiMn2O4 cell.

24

T. C. Hyams et al29 reported that the corrosion of aluminum current collector results in

battery’s power fade and capacity fade. The experiments were performed on 18650 cell with

LiNi0.8Co0.15Al0.5O2 as cathode and ethylene carbonate (EC) + ethylmethyl carbonate (EMC)

with 1.2 M LiPF6 as the electrolyte. Battery cycle-life tests were conducted at 25 ℃ for times

ranging from 4 weeks to 140 weeks. Also five cells were cycle-life tested at 45 ℃ for time

period ranging from 0 weeks to 68 weeks. A correlation between aluminum current collector

corrosion and battery performance fade was found, as listed in table 1.2. Corrosion of aluminum

is also confirmed by ICP tests, which shows around 2500 ppm of aluminum ions in electrolytes

after cycling for 52 weeks.

Table 1.2 Cycling parameters for tested cells at 25 ℃ and 45 ℃.

Cycling

time

Temp

℃

Corroded

area (%)

Pit density

(pits/cm2)

Power

fade (%)

Capacity

fade (%)

Al concentration

(ppm)

4 25 6.9 9.6×105 2 2.9

20 25 8.7 9.6×105 9.38 5.89

36 25 8.3 1.1×105 15.56 7.69

52 25 13.6 1.9×105 19.77 10.2 2529

68 25 10.9 1.3×105 25.28 11.55 2669

140 25 17.5 2.0×106 46.15 31.75

0 45 1.66 3.59×105 0 0

0 45 5.16 6.1×105 0 0

4 45 8.53 2.4×106 9.56 3.2

32 45 9.32 9.0×105 22.9 30.77

40 45 8.69 1.8×106 28.8 13.7

68 45 7.98 1.1×106 51.8 10.9

The results indicate that the corrosion is a significant cause of performance degradation.

There is a strong correlation between the fractions of area corroded and the power fade and

capacity fade of the tested batteries. Notably, corrosion is not only the factor that may cause

25

degradation of battery performance. Solid electrolyte interface deterioration, phase separation of

cathode and increased impedance could also be the reasons for capacity and power fade.

It is expected that the corrosion of aluminum current collector will impose at least three

adverse effects on the battery performance. First, the corrosion of aluminum will decrease the

effective interfacial area. Second, the corrosion products increase resistance of the interface of

cathode material and current collector. Last, the higher concentration of aluminum ion in

electrolyte may significantly impair the performance of cathode and anode.

1.5 Methods for protecting aluminum current collector from corrosion

1.5.1 Inhibitor

Physical adsorption and Chemisorption are two principle interactions between inhibitors

and the protected metal surface. Physical adsorption is due to the electrostatic attractive force

between inhibiting ions or dipoles and the electrically charged metal surface. The surface charge

on metal surface is ascribed to the electric field at the outer Helmhaltz plane of the electrical

double layer existing at metal/solution interface.36 The surface charge is defined as the potential

difference between Ecorr and zero charge potential (ZCP, Eq=0) of the metal. The adsorption

behavior is related to the charged compounds and the dipoles whose orientation is determined by

the surface charge. In chemisorption process, charge sharing or charge transfer happens from

inhibitor molecules to the metal surface.

Y. Li described a method of protecting aluminum current collector from corrosion in

electrolyte containing LiN(CF3SO2)2 (LITFSI) by using fumed silica nanoparticles.37,38 LiTFSI is

an appealing salt compared to LiPF6 and LiBF4 as it is more thermally and chemically stable.

However, it could not be applied in lithium ion battery because of its corrosive effect on

26

aluminum current collector. The fumed silica nanoparticles have hydrophilic silanol surface

groups, which makes the particles effective absorbents. It was presumed that the protection is

due to the enhanced adhesion of passive film or protective salt film to aluminum surface

compared to liquid without fumed silica. When fumed silica particles are blended into the

electrolyte to form gel electrolytes, the electrolytes exhibit desirable properties of both solid and

liquid, yet have high conductivity. The composite is shown to attenuate lithium dendrite growth

and improve charge-discharge performance. In open circuit potential measurements, the

Al/electrolyte interface shows more stability with the adding of fumed silica. The inhibiting

effect is attributed to higher viscosity of the composite electrolyte and it assist i